IT44458: amqmfsck or crtmqm might fail unexpectedly on NFS file system onAIX

APAR status

• Closed as program error.

Error description

• When testing a shared file system(NFS) on AIX the amqmfsck
  command might not work as expected and
  the crtmqm might fail with error 2195(MQRC_UNEXPECTED_ERROR)
  when creating the queue manage
  on the same file system.
  
  Output from amqmfsck during a successful test run:
  
  Node 1:
  ----------
  $ amqmfsck -wv /MQHA/qmgrs
  System call: stat("/MQHA/qmgrs",&statbuf)
  System call: statvfs("/MQHA/qmgrs")
  System call: fd =
  open("/MQHA/qmgrs/amqmfsck.lkw",O_EXCL|O_CREAT|O_RDWR,0666)
  System call: fchmod(fd,0666)
  System call: fstat(fd,&statbuf)
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  File lock acquired.
  System call: fork()
  System call: waitpid()
  Start a second copy of this program with the same parameters on
  another server.
  Press Enter or terminate this process to release the lock.
  
  System call: close(fd)
  File lock released.
  The tests on the directory completed successfully.
  ----------
  
  Node 2:
  ----------
  $  amqmfsck -wv /MQHA/qmgrs
  System call: stat("/MQHA/qmgrs",&statbuf)
  System call: statvfs("/MQHA/qmgrs")
  System call: fd =
  open("/MQHA/qmgrs/amqmfsck.lkw",O_EXCL|O_CREAT|O_RDWR,0666)
  System call: fchmod(fd,0666)
  System call: fstat(fd,&statbuf)
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  Waiting for the file lock.
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  Waiting for the file lock.
  ...
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  Waiting for the file lock.
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  File lock acquired.
  System call: fork()
  System call: waitpid()
  Press Enter or terminate this process to release the lock.
  
  System call: close(fd)
  File lock released.
  The tests on the directory completed successfully.
  ----------
  
  Output from amqmfsck during the failed test run:
  
  Node 1:
  ----------
  $ amqmfsck -wv /MQHA/qmgrs
  System call: stat("/MQHA/qmgrs",&statbuf)
  System call: statvfs("/MQHA/qmgrs")
  System call: fd =
  open("/MQHA/qmgrs/amqmfsck.lkw",O_EXCL|O_CREAT|O_RDWR,0666)
  System call: fchmod(fd,0666)
  System call: fstat(fd,&statbuf)
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  File lock acquired.
  System call: fork()
  System call: waitpid()
  ----------
  The tool did not display the message "Start a second copy of
  this program with the same parameters on another server.
  Press Enter or terminate this process to release the lock."
  
  Node 2:
  ----------
  $ amqmfsck -wv /MQHA/qmgrs
  System call: stat("/MQHA/qmgrs",&statbuf)
  System call: statvfs("/MQHA/qmgrs")
  System call: fd =
  open("/MQHA/qmgrs/amqmfsck.lkw",O_EXCL|O_CREAT|O_RDWR,0666)
  System call: fchmod(fd,0666)
  System call: fstat(fd,&statbuf)
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  Waiting for the file lock.
  System call: fcntl(fd,F_SETLK,F_WRLCK)
  System call: close(fd)
  File lock released.
  AMQ5667E: Error executing system call 'fcntl(F_SETLK)[w]' on
  file
  '/MQHA/qmgrs/amqmfsck.lkw', errno 22 ('xcsWaitFileLock()
  unexpected error').
  ----------
  When the amqmfsck instance ended on the Node1(e.g. using
  Ctrl+C), the amqmfsck
  waiting for the lock on Node1 ended with AMQ5667E.
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Users configuring MQ multi instance queue manager on AIX
  
  
  Platforms affected:
  AIX
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  A defect in MQ logic managing file locks caused the failure in
  amqmfsck and crtmqm.
  

Problem conclusion

• MQ code has been modified to correctly manage the file lock to
  resolve the failure in amqmfsck and crtmqm.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.3 LTS   9.3.0.20
  v9.x CD    9.3.3
  
  The latest available maintenance can be obtained from
  'IBM MQ Recommended Fixes'
  https://www.ibm.com/support/pages/recommended-fixes-ibm-mq
  
  If the maintenance level is not yet available information on
  its planned availability can be found in 'IBM MQ
  Planned Maintenance Release Dates'
  https://ibm.biz/mqplannedmaintenance
  
  ---------------------------------------------------------------
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  MQ BASE V9.3

• Fixed component ID

  5724H7291

Applicable component levels